Research map / 3D vision / Human pose estimation
Visibility Aware Human-Object Interaction Tracking From Single RGB Camera
Xianghui Xie, Bharat Lal Bhatnagar, Gerard Pons-Moll
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| SMPL-T conditioning of neural fields (vs fixed-depth human) (ablation) | SIF-Net joint tracking · BEHAVE sub-test | Object Chamfer, Align w=10 (cm): 17.29 -> 8.23 | Improved |
| HVOP-Net object pose prediction under occlusion (ablation) | full tracking pipeline · BEHAVE sub-test | Object Chamfer, Align w=10 (cm): 9.36 -> 8.23 | Improved |
| joint optimization (human, object, contact terms) (ablation) | full tracking pipeline · BEHAVE sub-test | Object Chamfer, Align w=10 (cm): 16.02 -> 8.23 | Improved |
| SMPL-T conditioning (ablation) | full tracking pipeline · BEHAVE sub-test | SMPL Chamfer, Align w=10 (cm): 14.40 -> 8.03 | Improved |
| HVOP-Net object pose prediction (vs CHORE-style single-frame) | object tracking on BEHAVE and InterCap · BEHAVE and InterCap (Align w=1 and w=10) | Object Chamfer (cm): BEHAVE w=10: 20.32 -> 8.49; InterCap w=10: 21.05 -> 11.38 | Improved |
| HVOP-Net vs SLERP/SmoothNet/CMIB infill alternatives (ablation) | object pose under occlusion · BEHAVE sub-test | rotation robustness under occlusion: qualitative/figure only, no numeric table | Improved |
| SmoothNet on SMPL-T parameters (ablation) | SMPL reconstruction · BEHAVE sub-test | Chamfer (cm): 8.71 -> 8.01 | Improved |
| SmoothNet on object pose before HVOP-Net input (ablation) | object pose estimation · BEHAVE sub-test | Chamfer (cm): Raw 5.03 -> Raw+SmoothNet 4.22 -> full 3.62 | Improved |
| SMPL-T conditioning for raw object prediction (ablation) | SIF-Net raw object pose · BEHAVE sub-test | Chamfer (cm), raw prediction: 5.56 -> 3.98 | Improved |
Closest papers by meaning
- Human-Aware Object Placement for Visual Environment Reconstruction (CVPR 2022 · similarity 0.89)
- ODHSR: Online Dense 3D Reconstruction of Humans and Scenes from Monocular Videos (CVPR 2025 · similarity 0.89)
- RHINO: Reconstructing Human Interactions with Novel Objects from Monocular Videos (CVPR 2026 · similarity 0.89)
- Seeing Behind Objects for 3D Multi-Object Tracking in RGB-D Sequences (CVPR 2021 · similarity 0.88)
- HOI-M^3: Capture Multiple Humans and Objects Interaction within Contextual Environment (CVPR 2024 · similarity 0.87)
- Reconstructing People, Places, and Cameras (CVPR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
